Qwen-Long - Alibaba Cloud Model Studio: Central de Documentação do Alibaba Cloud

O Qwen-Long processa documentos de até 10 milhões de tokens por meio de um mecanismo de upload e referência de arquivos, superando os limites de contexto dos modelos padrão.

ObservaçãoEste documento se aplica apenas à região da China continental (Beijing). Para usar o modelo, utilize uma chave de API da região da China continental (Beijing).

Como usar

Use o Qwen-Long em duas etapas: faça o upload dos arquivos e depois chame a API.

Upload e análise de arquivos:
- Faça o upload de um arquivo usando a API. Para obter detalhes sobre os formatos de arquivo suportados e limites de tamanho, consulte Formatos suportados.
- Após um upload bem-sucedido, o sistema retorna um file-id único para sua conta e inicia a análise. Não há cobrança pelo upload, armazenamento ou análise do arquivo.
Chamada de API e faturamento:
- Ao chamar o modelo, referencie um ou mais file-id s na mensagem system.
- O modelo executa a inferência com base no conteúdo de texto associado ao file-id.
- Em cada chamada de API, o número de tokens no conteúdo do arquivo referenciado conta como tokens de entrada daquela solicitação.

Essa abordagem evita a transferência de arquivos grandes em cada solicitação, mas observe que os tokens do arquivo são cobrados por chamada de API.

Primeiros passos

Pré-requisitos

Obtenha uma chave de API e configure-a como variável de ambiente.
Para chamar o modelo via SDK, instale o OpenAI SDK.

Fazer upload de um documento

Este exemplo faz o upload do arquivo Model_Studio_Phone_Product_Introduction.docx para o armazenamento seguro do Model Studio por meio da interface compatível com OpenAI e obtém um file-id. Consulte a documentação da API para ver os parâmetros de upload.

Python

import os
from pathlib import Path
from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),  # If not configured, replace with your API key.
    # The following URL is for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
    base_url="https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
)

file_object = client.files.create(file=Path("Model_Studio_Phone_Product_Introduction.docx"), purpose="file-extract")
print(file_object.id)

Java

import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.models.files.*;

import java.nio.file.Path;
import java.nio.file.Paths;

public class Main {
    public static void main(String[] args) {
        // Create a client and use the API key from the environment variable.
        OpenAIClient client = OpenAIOkHttpClient.builder()
                // If you have not configured the environment variable, replace the following line with: .apiKey("sk-xxx")
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                // The following URL is for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
                .baseUrl("https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1")
                .build();
        // Set the file path. Modify the path and filename as needed.
        Path filePath = Paths.get("src/main/java/org/example/Model_Studio_Phone_Product_Introduction.docx");
        // Create file upload parameters.
        FileCreateParams fileParams = FileCreateParams.builder()
                .file(filePath)
                .purpose(FilePurpose.of("file-extract"))
                .build();

        // Upload the file and print the file-id.
        FileObject fileObject = client.files().create(fileParams);
        System.out.println(fileObject.id());
    }
}

curl

# The following URL is for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
curl --location --request POST 'https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1/files' \
  --header "Authorization: Bearer $DASHSCOPE_API_KEY" \
  --form 'file=@"Alibaba Cloud Model Studio Phone Series Product Introduction.docx"' \
  --form 'purpose="file-extract"'

Execute o código para obter o file-id do arquivo enviado.

Passar informações e conversar usando um ID de arquivo

Passe o file-id nas mensagens do sistema: a primeira mensagem define a função, a segunda contém o file-id e, em seguida, adicione as perguntas do usuário.

Documentos mais longos exigem maior tempo de análise. Aguarde a conclusão da análise antes de fazer a chamada.

Python

import os
from openai import OpenAI, BadRequestError

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),  # If not configured, replace with your API key.
    # Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
    base_url="https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
)
try:
    # Initialize messages list.
    completion = client.chat.completions.create(
        model="qwen-long",
        messages=[
            # sys1: Role definition.
            {'role': 'system', 'content': 'You are a helpful assistant.'},
            # sys2: Document content (plain text or file-id).
            # Replace '{FILE_ID}' with the file-id used in your conversation.
            {'role': 'system', 'content': f'fileid://{FILE_ID}'},
            # When the request includes a second system message, the user message content is limited to 9,000 tokens.
            {'role': 'user', 'content': 'What is this article about?'}
        ],
        # All examples use streaming output to show the model's response process. For non-streaming examples, see https://www.alibabacloud.com/help/en/model-studio/text-generation
        stream=True,
        stream_options={"include_usage": True}
    )

    full_content = ""
    for chunk in completion:
        if chunk.choices and chunk.choices[0].delta.content:
            # Concatenate the output content.
            full_content += chunk.choices[0].delta.content
            print(chunk.model_dump())

        # Get token usage.
        if chunk.usage:
            print(f"Total tokens: {chunk.usage.total_tokens}")

    print(full_content)

except BadRequestError as e:
    print(f"Error: {e}")
    print("See documentation: https://www.alibabacloud.com/help/en/model-studio/error-code")

Java

import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.core.http.StreamResponse;
import com.openai.models.chat.completions.*;

public class Main {
    public static void main(String[] args) {
        // Create a client and use the API key from the environment variable.
        OpenAIClient client = OpenAIOkHttpClient.builder()
                // If you have not configured the environment variable, replace the following line with: .apiKey("sk-xxx")
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                // Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
                .baseUrl("https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1")
                .build();

        // Create a chat request.
        ChatCompletionCreateParams chatParams = ChatCompletionCreateParams.builder()
                //sys1: Role definition.
                .addSystemMessage("You are a helpful assistant.")
                //sys2: Document content (plain text or file-id).
                //Replace '{FILE_ID}' with the file-id used in your conversation.
                .addSystemMessage("fileid://{FILE_ID}")
                //When the request includes a second system message, the user message content is limited to 9,000 tokens.
                .addUserMessage("What is this article about?")
                .model("qwen-long")
                .build();

        StringBuilder fullResponse = new StringBuilder();

        // All examples use streaming output to show the model's response process. For non-streaming examples, see https://www.alibabacloud.com/help/en/model-studio/text-generation
        try (StreamResponse<ChatCompletionChunk> streamResponse = client.chat().completions().createStreaming(chatParams)) {
            streamResponse.stream().forEach(chunk -> {
                // Print and concatenate the content of each chunk.
                System.out.println(chunk);
                String content = chunk.choices().get(0).delta().content().orElse("");
                if (!content.isEmpty()) {
                    fullResponse.append(content);
                }
            });
            System.out.println(fullResponse);
        } catch (Exception e) {
            System.err.println("Error: " + e.getMessage());
            System.err.println("See documentation: https://www.alibabacloud.com/help/en/model-studio/error-code");
        }
    }
}

curl

# Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
curl --location 'https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1/chat/completions' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header "Content-Type: application/json" \
--data '{
    "model": "qwen-long",
    "messages": [
        {"role": "system","content": "You are a helpful assistant."},
        {"role": "system","content": "fileid://file-fe-xxx"},
        {"role": "user","content": "What is this article about?"}
    ],
    "stream": true,
    "stream_options": {
        "include_usage": true
    }
}'

Passar vários documentos

Passe vários file-id s em uma única mensagem do sistema ou adicione mensagens de sistema separadas para cada documento.

Pass multiple documents

Python

import os
from openai import OpenAI, BadRequestError

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),  # If not configured, replace with your API key.
    # Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
    base_url="https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
)
try:
    # Initialize messages list.
    completion = client.chat.completions.create(
        model="qwen-long",
        messages=[
            {'role': 'system', 'content': 'You are a helpful assistant.'},
            # Replace '{FILE_ID1}' and '{FILE_ID2}' with the file-ids used in your conversation.
            {'role': 'system', 'content': f"fileid://{FILE_ID1},fileid://{FILE_ID2}"},
            {'role': 'user', 'content': 'What are these articles about?'}
        ],
        # All examples use streaming output to show the model's response process. For non-streaming examples, see https://www.alibabacloud.com/help/en/model-studio/text-generation
        stream=True,
        stream_options={"include_usage": True}
    )

    full_content = ""
    for chunk in completion:
        if chunk.choices and chunk.choices[0].delta.content:
            # Concatenate the output content.
            full_content += chunk.choices[0].delta.content
            print(chunk.model_dump())

    print(full_content)

except BadRequestError as e:
    print(f"Error: {e}")
    print("See documentation: https://www.alibabacloud.com/help/en/model-studio/error-code")

Java

import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.core.http.StreamResponse;
import com.openai.models.chat.completions.*;

public class Main {
    public static void main(String[] args) {
        // Create a client and use the API key from the environment variable.
        OpenAIClient client = OpenAIOkHttpClient.builder()
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                // Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
                .baseUrl("https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1")
                .build();

        // Create a chat request.
        ChatCompletionCreateParams chatParams = ChatCompletionCreateParams.builder()
                .addSystemMessage("You are a helpful assistant.")
                //Replace '{FILE_ID1}' and '{FILE_ID2}' with the file-ids used in your conversation.
                .addSystemMessage("fileid://{FILE_ID1},fileid://{FILE_ID2}")
                .addUserMessage("What are these two articles about?")
                .model("qwen-long")
                .build();

        StringBuilder fullResponse = new StringBuilder();

        // All examples use streaming output to show the model's response process. For non-streaming examples, see https://www.alibabacloud.com/help/en/model-studio/text-generation
        try (StreamResponse<ChatCompletionChunk> streamResponse = client.chat().completions().createStreaming(chatParams)) {
            streamResponse.stream().forEach(chunk -> {
                // The content of each chunk.
                System.out.println(chunk);
                String content = chunk.choices().get(0).delta().content().orElse("");
                if (!content.isEmpty()) {
                    fullResponse.append(content);
                }
            });
            System.out.println("\nFull response content:");
            System.out.println(fullResponse);
        } catch (Exception e) {
            System.err.println("Error: " + e.getMessage());
            System.err.println("See documentation: https://www.alibabacloud.com/help/en/model-studio/error-code");
        }
    }
}

curl

# Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
curl --location 'https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1/chat/completions' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header "Content-Type: application/json" \
--data '{
    "model": "qwen-long",
    "messages": [
        {"role": "system","content": "You are a helpful assistant."},
        {"role": "system","content": "fileid://file-fe-xxx1"},
        {"role": "system","content": "fileid://file-fe-xxx2"},
        {"role": "user","content": "What are these two articles about?"}
    ],
    "stream": true,
    "stream_options": {
        "include_usage": true
    }
}'

Append documents

Python

import os
from openai import OpenAI, BadRequestError

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),  # If not configured, replace with your API key.
    # Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
    base_url="https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
)
# Initialize the messages list.
messages = [
    {'role': 'system', 'content': 'You are a helpful assistant.'},
    # Replace '{FILE_ID1}' with the file-id used in your conversation.
    {'role': 'system', 'content': f'fileid://{FILE_ID1}'},
    {'role': 'user', 'content': 'What is this article about?'}
]

try:
    # First-round response
    completion_1 = client.chat.completions.create(
        model="qwen-long",
        messages=messages,
        stream=False
    )
    # Print first-round response.
    # To stream: set stream=True, concatenate segments, and pass to assistant_message content.
    print(f"First-round response: {completion_1.choices[0].message.model_dump()}")
except BadRequestError as e:
    print(f"Error: {e}")
    print("See documentation: https://www.alibabacloud.com/help/en/model-studio/error-code")

# Construct the assistant_message.
assistant_message = {
    "role": "assistant",
    "content": completion_1.choices[0].message.content}

# Add assistant_message to messages.
messages.append(assistant_message)

# Add the file-id of the appended document to messages.
# Replace '{FILE_ID2}' with the file-id used in your conversation.
system_message = {'role': 'system', 'content': f'fileid://{FILE_ID2}'}
messages.append(system_message)

# Add the user's question.
messages.append({'role': 'user', 'content': 'What are the similarities and differences between the methods discussed in these two articles?'})

# Response after appending the document.
completion_2 = client.chat.completions.create(
    model="qwen-long",
    messages=messages,
    # All code examples use streaming output to clearly and intuitively show the model output process. For non-streaming output examples, see https://www.alibabacloud.com/help/en/model-studio/text-generation
    stream=True,
    stream_options={
        "include_usage": True
    }
)

# Stream and print the response after appending the document.
print("Response after appending the document:")
for chunk in completion_2:
    print(chunk.model_dump())

Java

import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.models.chat.completions.*;
import com.openai.core.http.StreamResponse;

import java.util.ArrayList;
import java.util.List;

public class Main {
    public static void main(String[] args) {
        OpenAIClient client = OpenAIOkHttpClient.builder()
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                // Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
                .baseUrl("https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1")
                .build();
        // Initialize messages list.
        List<ChatCompletionMessageParam> messages = new ArrayList<>();

        // Add information for role setting.
        ChatCompletionSystemMessageParam roleSet = ChatCompletionSystemMessageParam.builder()
                .content("You are a helpful assistant.")
                .build();
        messages.add(ChatCompletionMessageParam.ofSystem(roleSet));

        // Replace '{FILE_ID1}' with the file-id used in your conversation.
        ChatCompletionSystemMessageParam systemMsg1 = ChatCompletionSystemMessageParam.builder()
                .content("fileid://{FILE_ID1}")
                .build();
        messages.add(ChatCompletionMessageParam.ofSystem(systemMsg1));

        // User question message (USER role).
        ChatCompletionUserMessageParam userMsg1 = ChatCompletionUserMessageParam.builder()
                .content("Please summarize the article content.")
                .build();
        messages.add(ChatCompletionMessageParam.ofUser(userMsg1));

        // Construct the first-round request and handle exceptions.
        ChatCompletion completion1;
        try {
            completion1 = client.chat().completions().create(
                    ChatCompletionCreateParams.builder()
                            .model("qwen-long")
                            .messages(messages)
                            .build()
            );
        } catch (Exception e) {
            System.err.println("Request error. See error code page:");
            System.err.println("https://www.alibabacloud.com/help/en/model-studio/error-code");
            System.err.println("Error details: " + e.getMessage());
            e.printStackTrace();
            return;
        }

        // First-round response.
        String firstResponse = completion1 != null ? completion1.choices().get(0).message().content().orElse("") : "";
        System.out.println("First-round response: " + firstResponse);

        // Construct AssistantMessage.
        ChatCompletionAssistantMessageParam assistantMsg = ChatCompletionAssistantMessageParam.builder()
                .content(firstResponse)
                .build();
        messages.add(ChatCompletionMessageParam.ofAssistant(assistantMsg));

        // Replace '{FILE_ID2}' with the file-id used in your conversation.
        ChatCompletionSystemMessageParam systemMsg2 = ChatCompletionSystemMessageParam.builder()
                .content("fileid://{FILE_ID2}")
                .build();
        messages.add(ChatCompletionMessageParam.ofSystem(systemMsg2));

        // Second-round user question (USER role).
        ChatCompletionUserMessageParam userMsg2 = ChatCompletionUserMessageParam.builder()
                .content("Please compare the structural differences between the two articles.")
                .build();
        messages.add(ChatCompletionMessageParam.ofUser(userMsg2));

        // All examples use streaming output to show the model's response process. For non-streaming examples, see https://www.alibabacloud.com/help/en/model-studio/text-generation
        StringBuilder fullResponse = new StringBuilder();
        try (StreamResponse<ChatCompletionChunk> streamResponse = client.chat().completions().createStreaming(
                ChatCompletionCreateParams.builder()
                        .model("qwen-long")
                        .messages(messages)
                        .build())) {

            streamResponse.stream().forEach(chunk -> {
                String content = chunk.choices().get(0).delta().content().orElse("");
                if (!content.isEmpty()) {
                    fullResponse.append(content);
                }
            });
            System.out.println("\nFinal response:");
            System.out.println(fullResponse.toString().trim());
        } catch (Exception e) {
            System.err.println("Error: " + e.getMessage());
            System.err.println("See documentation: https://www.alibabacloud.com/help/en/model-studio/error-code");
        }
    }
}

curl

# Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
curl --location 'https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1/chat/completions' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header "Content-Type: application/json" \
--data '{
    "model": "qwen-long",
    "messages": [
        {"role": "system","content": "You are a helpful assistant."},
        {"role": "system","content": "fileid://file-fe-xxx1"},
        {"role": "user","content": "What is this article about?"},
        {"role": "system","content": "fileid://file-fe-xxx2"},
        {"role": "user","content": "What are the similarities and differences between the methods discussed in these two articles?"}
    ],
    "stream": true,
    "stream_options": {
        "include_usage": true
    }
}'

Passar informações como texto simples

Em vez de usar file-id s, passe o conteúdo do documento diretamente como string. Adicione as definições de função na primeira mensagem para evitar confusão com o conteúdo do documento.

Se o conteúdo do documento exceder 1 milhão de tokens, use um ID de arquivo devido aos limites de tamanho da API.

Simple example

Insira o conteúdo do documento diretamente na Mensagem do Sistema.

Python

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),  # Replace your API key here if you haven't set the environment variable
    # Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
    base_url="https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
)
# Initialize the messages list
completion = client.chat.completions.create(
    model="qwen-long",
    messages=[
        {'role': 'system', 'content': 'You are a helpful assistant.'},
        {'role': 'system', 'content': 'Alibaba Cloud Model Studio smartphone product introduction: Alibaba Cloud Model Studio X1 —————— Enjoy an ultimate visual experience: features a 6.7-inch 1440 x 3200 pixel ultra-clear screen...'},
        {'role': 'user', 'content': 'What does the article talk about?'}
    ],
    # All code examples use streaming output to clearly and intuitively show the model's output process. For non-streaming examples, see https://www.alibabacloud.com/help/en/model-studio/text-generation
    stream=True,
    stream_options={"include_usage": True}
)

full_content = ""
for chunk in completion:
    if chunk.choices and chunk.choices[0].delta.content:
        # Append output content
        full_content += chunk.choices[0].delta.content
        print(chunk.model_dump())

print(full_content)

Java

import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.core.http.StreamResponse;

import com.openai.models.chat.completions.*;

public class Main {
    public static void main(String[] args) {
        // Create a client using the API key from the environment variable
        OpenAIClient client = OpenAIOkHttpClient.builder()
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                // Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
                .baseUrl("https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1")
                .build();

        // Create a chat request
        ChatCompletionCreateParams chatParams = ChatCompletionCreateParams.builder()
                .addSystemMessage("You are a helpful assistant.")
                .addSystemMessage("Alibaba Cloud Model Studio smartphone product introduction: Alibaba Cloud Model Studio X1 —————— Enjoy an ultimate visual experience: features a 6.7-inch 1440 x 3200 pixel ultra-clear screen...")
                .addUserMessage("What does this article talk about?")
                .model("qwen-long")
                .build();

        StringBuilder fullResponse = new StringBuilder();

        // All examples use streaming output to show the model's response process. For non-streaming examples, see https://www.alibabacloud.com/help/en/model-studio/text-generation
        try (StreamResponse<ChatCompletionChunk> streamResponse = client.chat().completions().createStreaming(chatParams)) {
            streamResponse.stream().forEach(chunk -> {
                // Print and append each chunk's content
                System.out.println(chunk);
                String content = chunk.choices().get(0).delta().content().orElse("");
                if (!content.isEmpty()) {
                    fullResponse.append(content);
                }
            });
            System.out.println(fullResponse);
        } catch (Exception e) {
            System.err.println("Error: " + e.getMessage());
            System.err.println("For more information, see https://www.alibabacloud.com/help/en/model-studio/error-code");
        }
    }
}

curl

# Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
curl --location 'https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1/chat/completions' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header "Content-Type: application/json" \
--data '{
    "model": "qwen-long",
    "messages": [
        {"role": "system","content": "You are a helpful assistant."},
        {"role": "system","content": "Alibaba Cloud Model Studio X1 —— Enjoy an ultimate visual experience: features a 6.7-inch 1440 x 3200 pixel ultra-clear screen with a 120Hz refresh rate, ..."},
        {"role": "user","content": "What does this article talk about?"}
    ],
    "stream": true,
    "stream_options": {
        "include_usage": true
    }
}'

Pass multiple documents

Para passar vários documentos em um único turno de conversa, coloque o conteúdo de cada documento em uma Mensagem do Sistema separada.

Python

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),  # Replace your API key here if you haven't set the environment variable
    # Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
    base_url="https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
)
# Initialize the messages list
completion = client.chat.completions.create(
    model="qwen-long",
    messages=[
        {'role': 'system', 'content': 'You are a helpful assistant.'},
        {'role': 'system', 'content': 'Alibaba Cloud Model Studio X1————Enjoy an ultimate visual experience: features a 6.7-inch 1440 x 3200 pixel ultra-clear screen with a 120Hz refresh rate...'},
        {'role': 'system', 'content': 'Stardust S9 Pro —— A revolutionary visual feast: breakthrough 6.9-inch 1440 x 3088 pixel under-display camera design...'},
        {'role': 'user', 'content': 'What are the similarities and differences between the products discussed in these two articles?'}
    ],
    # All code examples use streaming output to clearly and intuitively show the model's output process. For non-streaming examples, see https://www.alibabacloud.com/help/en/model-studio/text-generation
    stream=True,
    stream_options={"include_usage": True}
)
full_content = ""
for chunk in completion:
    if chunk.choices and chunk.choices[0].delta.content:
        # Append output content
        full_content += chunk.choices[0].delta.content
        print(chunk.model_dump())

print(full_content)

Java

import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.core.http.StreamResponse;

import com.openai.models.chat.completions.*;

public class Main {
    public static void main(String[] args) {
        // Create a client using the API key from the environment variable
        OpenAIClient client = OpenAIOkHttpClient.builder()
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                // Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
                .baseUrl("https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1")
                .build();

        // Create a chat request
        ChatCompletionCreateParams chatParams = ChatCompletionCreateParams.builder()
                .addSystemMessage("You are a helpful assistant.")
                .addSystemMessage("Alibaba Cloud Model Studio smartphone product introduction: Alibaba Cloud Model Studio X1 —————— Enjoy an ultimate visual experience: features a 6.7-inch 1440 x 3200 pixel ultra-clear screen...")
                .addSystemMessage("Stardust S9 Pro —— A revolutionary visual feast: breakthrough 6.9-inch 1440 x 3088 pixel under-display camera design...")
                .addUserMessage("What are the similarities and differences between the products discussed in these two articles?")
                .model("qwen-long")
                .build();

        StringBuilder fullResponse = new StringBuilder();

        // All examples use streaming output to show the model's response process. For non-streaming examples, see https://www.alibabacloud.com/help/en/model-studio/text-generation
        try (StreamResponse<ChatCompletionChunk> streamResponse = client.chat().completions().createStreaming(chatParams)) {
            streamResponse.stream().forEach(chunk -> {
                // Print and append each chunk's content
                System.out.println(chunk);
                String content = chunk.choices().get(0).delta().content().orElse("");
                if (!content.isEmpty()) {
                    fullResponse.append(content);
                }
            });
            System.out.println(fullResponse);
        } catch (Exception e) {
            System.err.println("Error: " + e.getMessage());
            System.err.println("For more information, see https://www.alibabacloud.com/help/en/model-studio/error-code");
        }
    }
}

curl

# Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
curl --location 'https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1/chat/completions' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header "Content-Type: application/json" \
--data '{
    "model": "qwen-long",
    "messages": [
        {"role": "system","content": "You are a helpful assistant."},
        {"role": "system","content": "Alibaba Cloud Model Studio X1 —— Enjoy an ultimate visual experience: features a 6.7-inch 1440 x 3200 pixel ultra-clear screen with a 120Hz refresh rate..."},
        {"role": "system","content": "Stardust S9 Pro —— A revolutionary visual feast: breakthrough 6.9-inch 1440 x 3088 pixel under-display camera design..."},
        {"role": "user","content": "What are the similarities and differences between the products discussed in these two articles?"}
    ],
    "stream": true,
    "stream_options": {
        "include_usage": true
    }
}'

Append documents

Durante a interação com o modelo, talvez seja necessário adicionar novas informações de documentos. Para isso, anexe o novo conteúdo do documento como uma Mensagem do Sistema ao array Messages.

Python

import os
from openai import OpenAI, BadRequestError

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),  # Replace your API key here if you haven't set the environment variable
    # Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
    base_url="https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
)
# Initialize the messages list
messages = [
    {'role': 'system', 'content': 'You are a helpful assistant.'},
    {'role': 'system', 'content': 'Alibaba Cloud Model Studio X1 —— Enjoy an ultimate visual experience: features a 6.7-inch 1440 x 3200 pixel ultra-clear screen with a 120Hz refresh rate...'},
    {'role': 'user', 'content': 'What does this article talk about?'}
]

try:
    # First-round response
    completion_1 = client.chat.completions.create(
        model="qwen-long",
        messages=messages,
        stream=False
    )
    # Print the first-round response
    # For streaming output in the first round, set stream=True and concatenate each segment's content. Pass the concatenated string as the content when constructing assistant_message
    print(f"First-round response: {completion_1.choices[0].message.model_dump()}")
except BadRequestError as e:
    print(f"Error: {e}")
    print("For more information, see https://www.alibabacloud.com/help/en/model-studio/error-code")

# Construct assistant_message
assistant_message = {
    "role": "assistant",
    "content": completion_1.choices[0].message.content}

# Append assistant_message to messages
messages.append(assistant_message)
# Append new document content to messages
system_message = {
    'role': 'system',
    'content': 'Stardust S9 Pro —— A revolutionary visual feast: breakthrough 6.9-inch 1440 x 3088 pixel under-display camera design, delivering an immersive visual experience...'}
messages.append(system_message)

# Add user question
messages.append({
    'role': 'user',
    'content': 'What are the similarities and differences between the products discussed in these two articles?'
})

# Response after appending the document
completion_2 = client.chat.completions.create(
    model="qwen-long",
    messages=messages,
    # All code examples use streaming output to clearly and intuitively show the model's output process. For non-streaming examples, see https://www.alibabacloud.com/help/en/model-studio/text-generation
    stream=True,
    stream_options={"include_usage": True}
)

# Stream and print the response after appending the document
print("Response after appending the document:")
for chunk in completion_2:
    print(chunk.model_dump())

Java

import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.models.chat.completions.*;
import com.openai.core.http.StreamResponse;

import java.util.ArrayList;
import java.util.List;

public class Main {
    public static void main(String[] args) {
        OpenAIClient client = OpenAIOkHttpClient.builder()
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                // Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
                .baseUrl("https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1")
                .build();
        // Initialize the messages list
        List<ChatCompletionMessageParam> messages = new ArrayList<>();

        // Add role-setting information
        ChatCompletionSystemMessageParam roleSet = ChatCompletionSystemMessageParam.builder()
                .content("You are a helpful assistant.")
                .build();
        messages.add(ChatCompletionMessageParam.ofSystem(roleSet));

        // First-round content
        ChatCompletionSystemMessageParam systemMsg1 = ChatCompletionSystemMessageParam.builder()
                .content("Alibaba Cloud Model Studio X1 —— Enjoy an ultimate visual experience: features a 6.7-inch 1440 x 3200 pixel ultra-clear screen with a 120Hz refresh rate, 256GB storage, 12GB RAM, and a 5000mAh long-lasting battery...")
                .build();
        messages.add(ChatCompletionMessageParam.ofSystem(systemMsg1));

        // User question (USER role)
        ChatCompletionUserMessageParam userMsg1 = ChatCompletionUserMessageParam.builder()
                .content("Please summarize the article content")
                .build();
        messages.add(ChatCompletionMessageParam.ofUser(userMsg1));

        // Build the first-round request and handle exceptions
        ChatCompletion completion1;
        try {
            completion1 = client.chat().completions().create(
                    ChatCompletionCreateParams.builder()
                            .model("qwen-long")
                            .messages(messages)
                            .build()
            );
        } catch (Exception e) {
            System.err.println("Error: " + e.getMessage());
            System.err.println("For more information, see https://www.alibabacloud.com/help/en/model-studio/error-code");
            e.printStackTrace();
            return;
        }

        // First-round response
        String firstResponse = completion1 != null ? completion1.choices().get(0).message().content().orElse("") : "";
        System.out.println("First-round response: " + firstResponse);

        // Construct AssistantMessage
        ChatCompletionAssistantMessageParam assistantMsg = ChatCompletionAssistantMessageParam.builder()
                .content(firstResponse)
                .build();
        messages.add(ChatCompletionMessageParam.ofAssistant(assistantMsg));

        // Second-round content
        ChatCompletionSystemMessageParam systemMsg2 = ChatCompletionSystemMessageParam.builder()
                .content("Stardust S9 Pro —— A revolutionary visual feast: breakthrough 6.9-inch 1440 x 3088 pixel under-display camera design, delivering an immersive visual experience...")
                .build();
        messages.add(ChatCompletionMessageParam.ofSystem(systemMsg2));

        // Second-round user question (USER role)
        ChatCompletionUserMessageParam userMsg2 = ChatCompletionUserMessageParam.builder()
                .content("Please compare the structural differences between the two descriptions")
                .build();
        messages.add(ChatCompletionMessageParam.ofUser(userMsg2));

        // All examples use streaming output to show the model's response process. For non-streaming examples, see https://www.alibabacloud.com/help/en/model-studio/text-generation
        StringBuilder fullResponse = new StringBuilder();
        try (StreamResponse<ChatCompletionChunk> streamResponse = client.chat().completions().createStreaming(
                ChatCompletionCreateParams.builder()
                        .model("qwen-long")
                        .messages(messages)
                        .build())) {

            streamResponse.stream().forEach(chunk -> {
                String content = chunk.choices().get(0).delta().content().orElse("");
                if (!content.isEmpty()) {
                    fullResponse.append(content);
                }
            });
            System.out.println("\nFinal response:");
            System.out.println(fullResponse.toString().trim());
        } catch (Exception e) {
            System.err.println("Error: " + e.getMessage());
            System.err.println("For more information, see https://www.alibabacloud.com/help/en/model-studio/error-code");
        }
    }
}

curl

# Endpoint for the China (Beijing) region. Replace {WorkspaceId} with your Workspace ID. URLs vary by region.
curl --location 'https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1/chat/completions' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header "Content-Type: application/json" \
--data '{
    "model": "qwen-long",
    "messages": [
            {"role": "system","content": "You are a helpful assistant."},
            {"role": "system","content": "Alibaba Cloud Model Studio X1 —— Enjoy an ultimate visual experience: features a 6.7-inch 1440 x 3200 pixel ultra-clear screen with a 120Hz refresh rate..."},
            {"role": "user","content": "What does this article talk about?"},
            {"role": "system","content": "Stardust S9 Pro —— A revolutionary visual feast: breakthrough 6.9-inch 1440 x 3088 pixel under-display camera design, delivering an immersive visual experience..."},
            {"role": "user","content": "What are the similarities and differences between the products discussed in these two articles"}
        ],
    "stream": true,
    "stream_options": {
        "include_usage": true
    }
}'

Preços do modelo

China continental

Se você selecionar o escopo de implantação China continental, os recursos de computação para inferência do modelo ficarão restritos à China continental. Os dados estáticos são armazenados na região selecionada. Região suportada: China (Beijing).

Modelo	Versão	Janela de contexto	Entrada máx.	Saída máx.	Custo de entrada	Custo de saída	Cota gratuita Nota
Modelo	Versão	(tokens)			(por 1M tokens)		Cota gratuita Nota
qwen-long Chamadas em lote pela metade do preço	Estável	10.000.000	10.000.000	32.768	0,5 CNY	2 CNY	1 milhão de tokens cada Válido por 90 dias após ativar o Model Studio
qwen-long-latest Sempre corresponde à versão snapshot mais recente Chamadas em lote pela metade do preço	Mais recente				0,5 CNY	2 CNY
qwen-long-2025-01-25 Também conhecido como qwen-long-0125	Snapshot				0,5 CNY	2 CNY

Nome do modelo	Versão	Tamanho do contexto	Entrada máx.	Saída máx.	Custo de entrada	Custo de saída
qwen-long-latest Sempre possui os mesmos recursos da versão snapshot mais recente.	Mais recente	10.000.000	10.000.000	32.768	$0,072	$0,287
qwen-long-2025-01-25 Também conhecido como qwen-long-0125.	Snapshot

Nome do modelo

Versão

Tamanho do contexto

Entrada máx.

Saída máx.

Custo de entrada

Custo de saída

(Tokens)

(por 1 milhão de tokens)

qwen-long-latest

Sempre possui os mesmos recursos da versão snapshot mais recente.

Mais recente

10.000.000

32.768

$0,072

$0,287

qwen-long-2025-01-25

Também conhecido como qwen-long-0125.

Snapshot

Perguntas frequentes

O modelo Qwen-Long suporta o envio de jobs em lote?

Sim. O Qwen-Long suporta a API Batch da OpenAI com 50% das taxas de chamadas em tempo real. Envie jobs em lote como arquivos; os jobs são executados assincronicamente e retornam resultados após a conclusão ou timeout.
Onde os arquivos são salvos após o upload usando a API de arquivos compatível com OpenAI?

Os arquivos são enviados para o bucket do Model Studio sem custo. Consulte a API de Arquivos OpenAI para consultar e gerenciar arquivos.
O que é qwen-long-2025-01-25?

Trata-se de um snapshot de versão congelado em um ponto específico no tempo. É mais estável que latest e não tem data de expiração.
Como garantir que o modelo gere uma string JSON em formato padrão?

O qwen-long e todos os snapshots suportam saída estruturada. Especifique um JSON Schema para garantir um JSON válido que corresponda à sua estrutura.

Referência da API

Consulte Detalhes da API Qwen para ver os parâmetros de entrada e saída do modelo Qwen-Long.

Códigos de erro

Se a chamada do modelo falhar e retornar uma mensagem de erro, consulte Códigos de erro para resolução.

Limites

Dependências do SDK:
- Operações de arquivo (upload, exclusão, consulta) exigem um SDK compatível com OpenAI.
- Invoque modelos usando um SDK compatível com OpenAI ou Dashscope SDK.
Upload de arquivos:
- Formatos suportados: TXT, DOCX, PDF, XLSX, EPUB, MOBI, MD, CSV, JSON, BMP, PNG, JPG/JPEG e GIF.
- Tamanho do arquivo: O tamanho máximo para arquivos de imagem é 20 MB. Para outros formatos de arquivo, o limite é 150 MB.
- Cota da conta: Máximo de 10.000 arquivos ou 100 GB por conta. Os uploads falham quando qualquer um dos limites é atingido. Exclua arquivos para liberar cota. Consulte Compatível com OpenAI - Arquivo.
- Período de armazenamento: Atualmente, não há limite de expiração para arquivos armazenados.
Entradas da API:
- A primeira mensagem system define a função. A segunda contém o conteúdo do documento ou fileid://xxx. A mensagem user contém a consulta.
- Ao referenciar arquivos usando um file-id, uma única solicitação pode referenciar no máximo 100 arquivos.
- Com uma segunda mensagem system, o limite da mensagem user é de 9.000 tokens. Não há limite com apenas uma mensagem de sistema.
- O tamanho total do contexto é limitado a 10 milhões de tokens.
Saídas da API:
- O comprimento máximo de saída é 32.768 tokens.
Compartilhamento de arquivos:
- Os file-id s são específicos da conta e não podem ser usados entre contas diferentes ou com chaves de API de usuários RAM.
Limitação de taxa: Para obter informações sobre as condições de limitação de taxa do modelo, consulte Limitação de taxa.